2010年10月05日

書評

またもや、全然更新してませんでした。9月に博士論文の諮問を受け、一部修正後、先週大学に提出しました。これで、博士取得ということになります。そのうち、5年間の博士取得までの道のりをもとめる予定です。

今回の記事は、科学哲学最新号に載った認識論関係の論文の書評、というか批判的コメントです。もとの論文が英語だったので、英語で書きました。

I've read "On Memory Knowledge" by Shin Sakuragi (Kagaku Tetsugaku 43-1, 2010, pp. 61-77). It is rare that a Japanese writes an epistemology paper, and so I will comment on this paper.


1. The author starts the paper by mentioning the problem of forgotten evidence, though, oddly, no reference to Goldman, who first raised this problem.

2. The central concept, memory impression, is not defined well. It is defined to be phenomenal, but it is often treated as having propositional contents.

3. on p. 63, internalism is characterized in terms of accessibility. Then, later, Conee & Feldman are discussed as proponents of internalism. But, of course, they don't commit themselves to access-internalism. For this reason, I don't think the author's interpretation of their view on p. 65 is right. Indeed, the quote there suggests that they advocate a different view - mentalism.

4. on p. 64, the author says "It is well recognized that we cannot base memory knowledge on any inferential process from memory impressions." It's not clear at all what this means, nor is why the author's argument shows this.

His argument is the following:

"to deduce the memory knowledge from my memory impression, I need to appeal to a principle like this: when I feel just like how I am feeling now, I almost always remember, and thereby know, something... But I cannot derive the principle by induction, for, to derive it inductively, I have to appeal to different instances of my past experience (“When I felt such and such at t, I actually remembered. . .”). On the face of it, such an appeal eventually leads me to a vicious regress."

I have no idea why this is relevant. First, there is a level-confusion in Alston's sense. In this paper in general, the author ignores important epistemic distinctions, such as "justification-knowledge," "belief-knowledge," and "propositional-doxastic justification." For example, when he claims "memory impression justifies memory knowledge," it means that memory impression knowledge-level justifies p. Relevant here is the distinction between first-order and second-order justification (or knowledge.) In the context where the quote is located, the issue is whether one can base p, not knowing p, on the memory impression that it seems that p. But then, the issue is changed to the one as to whether one can base knowing p.

Second, even if the issue is second-order justification or knowledge, I don't see why the argument holds. Though I'm not entirely sure, it seems that the author's point is that in order to have knowledge that S knows p on the basis of memory impression, S needs to have ground for the reliability of memory. But, in order to have ground for the reliability of memory, S needs to have ground for remembering p, which requires having ground for the reliability of memory, and so on (I don't know if this is a vicious regress; it is more of a circularity). But grounds for the reliability of memory can be gained from some source independent of memory, say, someone's testimony. Or, his point might be that this inference is bootstrapping. If so, it is a general problem for basic sources, arising for both internalism and externalism.

Third, this argument presupposes that what justifies must be justified all the way down. It is a Pyrrhonian skeptical argument. Even hard-core internalists would deny this presupposition.

5. on p. 69, Huemer's example is mentioned. I doubt if this is a good example for semantic reasons, but let me put this point aside. After all, it is simply a variant of the new evil demon case. Then, the apparent conclusion would be that internalist coherence is sufficient for justification, not that memory impressions justify beliefs, as the author derives it.

6. The example on pp. 70-1 does not show that "my memory impression constitutes a part of the epistemic grounds of my memory knowledge" (p. 71). It is consistent with the example that the defeater undermines the belief that S's memory is reliable, and it is why S is not justified on the basis of memory (not necessarily, memory impression). The concept of defeater has wider usage than Pollock construes it to be. Moreover, the argument here presupposes a foundationalist version of the present ground theory, according to which memory beliefs are ultimately grounded on memory impressions. it's odd that coherentism is not even mentioned, and the present ground theory is always assumed to be foundationalist.

7. I don't buy the counterexample to reliabilism on p. 71. I do not share the author's intuition, and the example doesn't involve any explanation of why his intuition seems right. Moreover, the belief in question (My name is X) is similar to necessary truth. Reliabilism has notorious difficulties with such truth. The problem doesn't seem specific to memory knowledge.

8. Again, I don't buy the example on p, 72-3. Especially, reliablism has an easy way-out of this example: memory in amnesia is a different process than memory in normal condition.

9. Many issued brought up in this paper are simple variants of the issues concerning perceptual knowledge. I wonder if there is any problem specific to memory knowledge other than the problem of forgotten evidence.
posted by hakutaku at 20:39| Comment(0) | TrackBack(0) | 認識論 | このブログの読者になる | 更新情報をチェックする

広告


この広告は60日以上更新がないブログに表示がされております。

以下のいずれかの方法で非表示にすることが可能です。

・記事の投稿、編集をおこなう
・マイブログの【設定】 > 【広告設定】 より、「60日間更新が無い場合」 の 「広告を表示しない」にチェックを入れて保存する。


×

この広告は1年以上新しい記事の投稿がないブログに表示されております。